It has long been known that slight chemical modifications of the morphine molecule lead to analgesic agonists of widely differing potency and addictive properties. For example, codeine, the methyl ether of morphine, is a relatively mild analgesic agonist having slight dependance (addiction) liability. On the other hand, heroin, the diacetyl derivative of morphine, is a powerful agonist with extremely high addiction potential. In addition, as long ago as 1915, Pohl found that when the N-methyl group of codeine was replaced with an allyl group, the resulting compound, N-allylnorcodeine, was an opiate antagonist. In 1940, N-allylnormorphine or nalorphine was synthesized and was shown to have a highly specific ability to reverse the depressant effects of morphine. Other simple chemical modifications of the morphine molecule have yielded many interesting drugs. Thus, one fruitful research area in the search for improved analgesics of high potency and/or lower dependance (addiction) liability has been the chemical modification of the morphine molecule.
In addition to modifying the morphine ring structure by chemical means, chemists have developed a second related field of research--the preparation of certain morphine part-structures--with the same end in mind as above; i.e., the synthesis of improved analgesic agonists and/or analgesic antagonists of improved properties. For example, meperidine, a widely used analgesic, can be written as a morphine part-structure. Many other morphine part-structures have been prepared, some of which have improved analgesic agonist properties and others, particularly those with an allyl group attached to a ring nitrogen, have opiate antagonist properties. It had been hoped that morphine part-structure research would produce a compound having both opiate agonist and antagonist properties since the opiate antagonist property would assure a user that the compound would have a greatly reduced dependance liability. Two recently marketed analgesics, pentazocine and phenazocine, have been found to be both antagonists and agonists although they still retain a certain degree of opiate dependance liability.
One potential morphine part-structure can be written as a decahydroisoquinoline with an hydroxyphenyl group substituted on a ring junction carbon atom para to the isoquinoline nitrogen. An attempt to prepare such a compound was described by Boekelheide in a paper appearing in J. Am. Chem. Soc., 69, 790 (1947). This paper set forth the preparation of what, according to the numbering system then in vogue, were 10-phenyldecahydroisoquinolines. It was the author's conclusion, however, that the compound (IX) had a cis configuration and (footnote 5) showed low analgesic activity. The synthesis itself is cumbersome and not free from ambiguity. Sugimoto et. al., J. Pharm. Soc. Japan, 75 177 (1955), C.A. 1956 1814b described the synthesis of 8 or 10-alkylated decahydroquinolines. The reference also shows the morphine part-structure, 10-(m-hydroxyphenyl)-3-methylisoquinoline [presently named as 1-methyl-3a-(m-hydroxyphenyl)-1,2,3,3a,4,5,6,7,7a,8-decahydroisoquinoline] but without furnishing a synthesis for it. These authors do not, in fact, describe the preparation of any decahydroisoquinoline, but describe only the preparation of the decahydroquinoline analogs.
Belgian Pat. No. 802,557 issued Jan. 19, 1974, discloses a general method of preparing N-substituted 3a-phenyldecahydroisoquinolines and specifically discloses 3a-phenyl,3a-(m-methoxy phenyl) and 3a-(m-hydroxyphenyl)-1-methyldecahydroisoquinolines, 3a-(m-methoxyphenyl) and 3a-(m-hydroxyphenyl)-1-phenethyldecahydroisoquinolines, and 1-cyclohexylmethyl-3a-phenyldecahydroisoquinoline.